Our preliminary studies of gene dosage abnormalities in ovarian cancer reveal numerous recurring abnormalities, many not previously known. WE BELIEVE THAT MANY OF THESE ABNORMALITIES REPRESENT GENETIC CHANGES THAT ENABLE TUMOR PROGRESSION AND THAT IDENTIFICATION OF THE INVOLVED GENES WILL LEAD TO IMPROVED DIAGNOSIS, PROGNOSIS AND THERAPY. The objectives of this proposal are to carefully map the regions of common abnormality, to investigate the role of these abnormalities in ovarian cancer progression and to begin the process of identifying the involved genes. We will accomplish these through the combined analysis of loss of heterozygosity (LOH), comparative genomic hybridization (CGH) and fluorescence in situ hybridization (FISH). Our specific aims are to; 1. Precisely define regions of gene dosage abnormality in serous grade III tumors and explore the chromosomal events that are involved. The number of grade III tumors analyzed using CGH, FSH and LOH analyses of primary tumors will be increased to 100 in order to allow statistically precise characterization of abnormalities that occur at frequencies greater than 10% These combined data will be analyzed at each site of gene dosage abnormality to determine whether the abnormality occurred as a result of physical deletion, mitotic recombination, loss plus reduplication, development of aneuploidy, amplification, etc. In addition, FISH with probes to regions of amplification identified using CGH will be applied to metaphase spreads in short term cultures, when available, to determine whether amplification is extrachromosomal or intrachromosomal. We also will compare our CGH results with CGH measurements of 100 matched tumors made by Japanese collaborators in order to investigate possible ethnic or environmental differences between ovarian cancer in Japan and the United States. 2. Define the order in which gene dosage abnormalities occur during progression. Our preliminary analyses of ovarian cancers show that high grade tumors have accumulated numerous abnormalities, many highly correlated. We presume that these abnormalities contribute to progression and we will test the hypothesis that they occur more or less serially by analysis of 100 borderline + grade I tumors and 100 grade II tumors. We will explore the possibility that correlations between abnormalities provide additional information about athe nature of the progression process. 3. Begin analysis of how the gene dosage abnormalities influence biological and clinical behavior by identifying genes associated with regions of common gene dosage abnormality in order to understand their genetic nature and biological function. Work during this project period will focus on 3q26 and 16q and on genes under study in other projects in this Program Project. Specific genes to be studied include EGF, EGFR, TGFalpha, TGFbeta, VEGF, VEGFR, KGF, KGFR, HGF and HGFR.